http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51529
--- Comment #3 from paul.richard.thomas at gmail dot com <paul.richard.thomas at gmail dot com> 2011-12-14 08:55:08 UTC --- Dear Tobias, > However, I do now understand why one needs for SOURCE= to memset the source to > NULL - at least as long _copy not only copies the data but also frees it. The > latter could be also left to _free. - Actually, I am in favour of separating > _copy and _free. As this issue shows, there are cases where one does not want > to combine them, leading to work around actions (memset). I think only for > polymorphic assignment, one needs _free + _copy, for allocate with SOURCE= a > _copy should be enough. The memset came about for similar reasons with class objects with allocatable components. I had missed this wrinkle with the testcase. My inclination is to restire the memset and keep the PR open. I am am trying to clear up some issues of functionality, starting with the failure of a%disp() to scalarize properly in class_array_3.f03. Then I have mind to understand the failure of vector indexing in gfc_trans_call and finally to deal with class array components and class array subreferences. After the tidy up :-) Cheers Paul
